|
Affiliation
|
Department of Orthopaedic Surgery, William Beaumont Army Medical Center, Texas Tech University Health Sciences Center, 5005 N. Piedras St, El Paso, TX 79920, USA. Electronic address: ajschoen@neomed.edu. |
|
Abstract
|
BACKGROUND CONTEXT: Predictors of complications and mortality after spine trauma are underexplored. At present, no study exists capable of predicting the impact of demographic factors, injury-specific predictors, race, ethnicity, and insurance status on morbidity and mortality after spine trauma.
PURPOSE: This study endeavored to describe the impact of patient demographics, comorbidities, injury-specific factors, race/ethnicity, and insurance status on outcomes after spinal trauma using the National Sample Program (NSP) of the National Trauma Data Bank (NTDB). STUDY DESIGN: The weighted sample of 75,351 incidents of spine trauma in the NTDB was used to develop a predictive model for important factors associated with mortality, postinjury complications, length of hospital stay, intensive care unit (ICU) days, and time on a ventilator. PATIENT SAMPLE: A weighted sample of 75,351 incidents of spine trauma as contained in the NTDB. OUTCOME MEASURES: Mortality, postinjury complications, length of hospital stay, ICU days, and time on a ventilator as reported in the NTDB.
METHODS: The 2008 NSP of the NTDB was queried to identify patients sustaining spine trauma. Patient demographics, race/ethnicity, insurance status, comorbidities, injury-specific factors, and outcomes were recorded, and a national estimate model was derived. Unadjusted differences in baseline characteristics between racial/ethnic groups and insurance status were evaluated using the t test for continuous variables and Wald chi-square analysis for categorical variables with Bonferroni correction for multiple comparisons. Weighted logistic regression was performed for categorical variables (mortality and risk of one or more complications), and weighted multiple linear regression analysis was used for continuous variables (length of hospital stay, ICU days, and ventilator time). Initial determinations were checked against a sensitivity analysis using imputed data.
RESULTS: The weighted sample contained 75,351 incidents of spine trauma. The average age was 45.8 years. Sixty-four percent of the population was male, 9% was black/African American, 38% possessed private/commercial insurance, and 12.5% lacked insurance. The mortality rate was 6% and 16% sustained complications. Increased age, male gender, Injury Severity Score (ISS), and blood pressure at presentation were significant predictors of mortality, whereas age, male gender, other mechanism of injury, ISS, and blood pressure at presentation influenced the risk of one or more complications. Nonwhite and black/African American race increased risk of mortality, and lack of insurance increased mortality and decreased the number of hospital days, ICU days, and ventilator time.
CONCLUSIONS: This is the first study to postulate predictors of morbidity and mortality after spinal trauma in a national model. Race/ethnicity and insurance status appear to be associated with greater risk of mortality after spine trauma.
Language: en |